RCUS - NYSE

Hello, everyone, and thank you for joining the Arcus Biosciences Full Year and Fourth Quarter 2025 Earnings and Financial Results Call. My name is Claire and I'll be coordinating your call today. [Operator Instructions] I will now hand over to Holli Kolkey from Arcus Biosciences to begin. Please go ahead.

Good afternoon and thank you for joining us on today's conference call to discuss Arcus' fourth quarter and full year 2025 financial results and pipeline updates. I will be filling in for Pia Eaves, our Head of IR, who is out on maternity leave. I would like to remind you that on this call, management will make forward-looking statements, including statements about our cash runway, our projected 2026 revenue and our expected clinical development milestones and timelines. All statements other than historical facts reflect the current beliefs and expectations of management and involve risks and uncertainties that may cause our actual results to differ from those expressed. Those risks and uncertainties are described in our most recent annual report on Form 10-K that has been filed with the SEC. For today's call, please refer to our latest corporate presentation posted in the Investors section of our website. This afternoon, you'll hear from several members of our management team. So now I'll turn the call over to our CEO, Terry Rosen, to begin.

Thanks very much, Holli, and thank you, everyone, for joining us this afternoon. 2026 is going to be a transformative year for Arcus. As you know, we are -- we've been focused on establishing casdatifan as the unequivocal best-in-class HIF-2 alpha inhibitor and the new standard of care for clear cell renal cell carcinoma. And this year is going to be another substantial year for data presentations as well as the advancement and expansion of our Phase III clinical program for CAS. I want to emphasize, particularly those who are new to Arcus or casdatifan that the advantages of casdatifan are well understood. That's all connect from the earliest days of its design and development, the advantages derived from dramatic differentiation of casdatifan's PK/PD profile. These are evident in the highly quantitative and reproducible differentiation on the primary biomarker for HIF-2 alpha inhibition, EPO production and the manifestation is improvement on all key efficacy measures. Casdatifan hits the target harder, hits it earlier. Just this week, we shared updated data from our ARC-20 cohorts evaluating casdatifan monotherapy in late-line clear cell RCC. We're also thrilled that Dr. Toni Choueiri will be presenting these data this weekend at ASCO GU. We're going to discuss these data in more detail today, but suffice it to say, the bottom line is that single-agent CAS continues to achieve unprecedented ORR and PFS in late-line clear cell RCC. It's not only relative to data for belzutifan, the only currently marketed HIF-2 alpha inhibitor, but also relative to data for standard of care TKIs. This can be seen very clearly on Slide 7 of our corporate deck. Importantly, casdatifan achieves these outcomes without the debilitating toxicities associated with TKIs. In addition to the clinical data, the ASCO GU presentation will include biomarker data that further reinforce the confidence in the differentiation of casdatifan versus belzutifan. Also at ASCO GU, we're going to see the detailed results from the Phase III LITESPARK-0011 (sic) [ LITESPARK-011 ] study. This evaluated belzutifan plus lenvatinib versus cabozantinib in IO-experienced clear cell RCC. These data should be both validating and highly derisking for our ongoing Phase III PEAK-1 study, which is evaluating cas plus cabo in a similar setting and with the same control arm. With both our own data and the belzutifan data being presented, this ASCO GU will be an extremely important event for the HIF-2 alpha inhibitor class, firmly establishing it as a key standard of care in the treatment of RCC. We believe HIF-2 alpha inhibitors will have a place in every line of treatment for RCC, and CAS is extremely well positioned to be the HIF-2 alpha inhibitor of choice across all settings. In fact, as we will describe later, we believe the profile of CAS will enable a unique frontline regimen that will transform the patient journey in the setting and could translate into multibillion-dollar commercial opportunity. Our first Phase III study for CAS PEAK-1 is enrolling and is designed to get CAS approved and to patients as quickly as possible. This represents our fast-to-market strategy. There is already a high level of excitement driving enrollment in PEAK-1 and the strength of our new ARC-20 data, coupled with the further validation of the HIF-2 alpha inhibition in early line settings by LITESPARK data, will amplify the enthusiasm for the study. So by combining our best-in-class HIF-2 alpha inhibitor with the most widely used TKI cabozantinib, we believe we'll capture substantial share of the IO experience setting. Now I'd like to spend a few minutes on our frontline strategy because this is going to be a huge focus for us throughout 2026. Our frontline strategy is enabled by the consistently low rate of primary progression that's been observed with casdatifan across settings. This is shown very clearly on Slide 20 of our corporate presentation. Primary progression reflects the proportion of patients whose disease progresses at or before the first scan. It's important for this rate to be as low as possible, particularly in early line treatments because patients with primary progression do not get an opportunity to benefit from therapy. This is devastating for both patients and their doctors. In contrast to the low rates for CAS, belzutifan is associated with a very high rate of primary progression, 35% is monotherapy in its Phase III trial. The manifestation of this key differentiation is that in frontline RCC, belzutifan will likely always require combination with a TKI to keep primary progression low. In fact, Merck's Phase III study in the frontline setting is evaluating exactly that. Lenva, belz, pembro -- that's a pretty nonpatient-friendly regimen in the context of quality of life. A TKI-free regimen, on the other hand, is much more desirable for both patients and clinicians. The most common feedback that we received from investigators is that given casdatifan's profile, the use of a TKI can likely be put off for years. This offers a far better option for patients that would greatly improve their quality of life. Therefore, our frontline strategy is to develop casdatifan without a TKI and specifically with a backbone of casdatifan plus anti-PD-1, which we can build upon with a third non-TKI mechanism. We plan to execute on this strategy quickly and efficiently using our ARC-20 study. It's probably a good time to explain how we have and will continue to leverage ARC-20 to drive our development strategy for CAS. First, with 4 monotherapy cohorts and 121 patients of efficacy data in late-line clear cell RCC, ARC-20 enabled us to clearly demonstrate that CAS has the best-in-class HIF-2 alpha inhibitor profile. Second, with these 4 monotherapy cohorts, which were designed to satisfy Project Optimus, we've established that 100 milligrams once a day is the optimal going-forward dose of casdatifan. Finally, the design allows us to rapidly and efficiently add and enroll cohorts to evaluate CAS and CAS-based combinations in other settings. We now have around 30 sites across 4 countries active in the study, and this drives efficiency. We first utilized this with the cas plus cabo cohort where we quickly generated data to support our first Phase III study, PEAK-1. We then added 3 new cohorts, approximately 90 patients in total, to demonstrate the feasibility of using CAS without a TKI in early line settings. One of these cohorts is the cas plus zim cohort, which is fully enrolled and for which we've already shared a primary progression rate of 9% for the first 23 of 30 patients. With the low rate of primary progression across all settings, we and our investigator advisers are convinced that the ideal frontline therapy is a TKI-sparing casdatifan regimen. And we just started enrolling a new cohort to evaluate CAS plus anti-PD-1 and anti-CTLA-4 to support the rapid initiation and execution of our first Phase III study in the frontline setting. Finally, while RCC is our top priority, we've generated exciting preclinical data for CAS and HCC, and are evaluating opportunities to pursue HCC in a cost and resource-efficient manner. We spent a lot of time already on casdatifan, but I want to transition now to our immunology portfolio, where there's been a lot of and growing interest. We've leveraged the same small molecule capability that created casdatifan to build an emerging portfolio of inflammation and immunology programs. Two of these are expected to enter the clinic over the next 12 months. We are focused on addressing validated targets against which it has historically been difficult to create small molecule drugs that have optimal pharmaceutical properties. For this reason, we expect limited competition for our I&I programs, similar to what we're seeing with casdatifan. Our 3 most advanced molecules are an MRGPRX2 antagonist, a TNF inhibitor, and CCR6 antagonist. Later on this call, Juan will speak in more detail about the potential differentiation of our compound relative to others. But before we go there, I'd like to turn the call over to Richard to review the new and updated casdatifan data that we'll be presenting at ASCO GU this coming weekend.

Thank you, Terry. I'd like to start on Slide 9 of our corporate deck, just to remind everyone about the design of our ARC-20 study. Today's data include updated ORR and PFS based on a data cutoff date of January 30 from the 4 late-line monotherapy cohorts highlighted on this slide. This is now the fourth time we are presenting data for single-agent casdatifan in a setting. And as you'll see on the next few slides, the efficacy data continued to improve with longer follow-up. Now moving to Slide 10, where we show the latest ORRs for the 100-milligram QD cohort, which is our going-forward dose and formulation. The confirmed ORR increased from 35% at the August data cut now to 45%. The 45% ORR in this late-line patient population is remarkable and that it's twice that observed with belzutifan in LITESPARK-5 (sic) [ LITESPARK-005 ]. Similarly, the confirmed ORR for the pooled analysis improved from 31% to 35%, well above the range of ORRs that have been observed for belzutifan. On Slide 12, we show the latest Kaplan-Meier curve for the 100-milligram cohort. And you can see here that this 100-milligram cohort shows an impressive median PFS of 15.1 months after 17.8 months of median follow-up. There are some patients still on treatment who are censored before the median, but even in a highly unlikely scenario where all censored patients progress at the next scan, the median PFS for this cohort would still be 14.4 months. Let's move on to the next slide, Slide 13, and we show the latest Kaplan-Meier curve for the pooled analysis. The median PFS remained at 12.2 months. And as you can see here, there are quite a lot of patients still on treatment beyond 12 months and even beyond 24 months. So overall, we are seeing PFS that is 2 to 3x longer with CAS monotherapy than the 5.6 months observed for belzutifan in the same setting. These data clearly support the proposition that casdatifan is the best-in-class HIF-2 alpha inhibitor, and our highest priority now is to maximize the potential of this molecule in clear cell renal cell carcinoma. So I'd now like to spend some time on our development plans, starting with our first Phase III study, PEAK-1, which is evaluating cas plus cabo versus cabo in immune therapy experienced ccRCC. The trial design is shown on Slide 19 of our corporate deck. PEAK-1 is actively enrolling, and we are confident that the study will complete enrollment quickly. Additionally, PEAK-1 has a sole primary endpoint of PFS, which should enable a relatively short time to read out. So this is a fast-to-market strategy that builds on top of standard of care. And based on all the data to date, we have high confidence that this study will establish cas plus cabo as the new standard of care in IO-experienced clear cell RCC. Meanwhile, as Terry talked about earlier, our first-line strategy has the potential to transform the treatment paradigm for RCC. We're focused on the development of a TKI-free regimen that improves on the efficacy of IO-only therapy, and there are several opportunities for CAS to do this. First, the biggest limitation of anti-PD-1 plus anti-CTLA-4 therapy is that it has a relatively high rate of primary progression that is of 20% to 25%. With CAS's low rate of primary progression and its orthogonal mechanism, we believe we can meaningfully improve upon this high rate of primary PD seen with the IO-only therapy. In fact, we have shown that patients who progress quickly on IO-based therapies have tumor markers that correlate with responsivity to CAS. Additionally, median PFS for IO-only regimens in clear cell RCC is relatively short at about 12 months. So again, we believe we can meaningfully improve on this PFS. To determine the optimal go-forward TKI-free regimen, we are evaluating multiple CAS combinations in ARC-20. These include our fully enrolled cohort evaluating cas plus zim, which we believe will convincingly demonstrate the ability of CAS plus an anti-PD-1 to form the backbone of our first-line regimen based on both safety and efficacy. We plan to share data from this cohort in the second half of this year. Additionally, we just started a cohort to evaluate CAS in combination with anti-PD-1 and anti-CTLA-4. In addition to this cohort, we also plan to evaluate another TKI-free combination for CAS in the frontline setting, and we'll share more about this in coming months. Lastly, we continue to monitor data from the EVOLVE study, evaluating CAS plus forimtamig, an anti-PD-1 CTLA-4 bispecific that's in collaboration with Astra

Thanks, Richard. I'd like to start on Slide 22. RCC represents a massive multibillion-dollar market opportunity for casdatifan. Sales for RCC drugs in just some major markets are over $10 billion annually today and anticipated to grow to $13 billion by 2030. As you can see here, historically the market has been dominated by only 2 classes of therapies, IO and TKIs, and HIF-2 alpha inhibition is the only new class of drugs on the horizon today. Importantly, as of today, the HIF-2 alpha inhibitor field is only a 2-horse race with just belzutifan and us with casdatifan. While Merck is slightly ahead of us, historical data has shown that in just a 2-player market in oncology, second entrants have captured 42% when there is 0 differentiation. And let me repeat that, this is when there is no differentiation. With efficacy differentiation as we are seeing with casdatifan, this share can be meaningfully higher and oncology analogs have shown that fast followers with differentiation share can be as high as 85%. Additionally, we are developing CAS with different combination partners of belzutifan, which should drive even further differentiation. Turning to Slide 23. The sole marketed HIF-2 alpha inhibitor, belzutifan, which is currently approved only in late-line clear cell RCC is already generating annual run rate sales of nearly $1 billion. While impressive, this is only scratching the surface in terms of market opportunity for a best-in-class HIF-2 alpha inhibitor. For casdatifan, we are focused on early line settings, which have larger patient populations and longer durations of therapy, both of which will contribute to a much larger market opportunity that is multibillion dollars in size. Specifically, our PEAK-1 study is targeting the IO experience setting, of which there are approximately 21,000 patients in the major markets alone, so more than twice the third-line patient population where belz is approved today. Additionally, with the duration of therapy at least double that of belzutifan today, we expect a peak sales opportunity of $2.5 billion in the IO experienced or PEAK-1 setting alone. In first line, the opportunity is even greater. With our TKI-free strategy, we believe we can take meaningful share for both IO/IO and IO/TKI regimens. Here, we estimate peak sales potentially for CAS of $3 billion or more. All in, casdatifan could be a $5 billion drug in first- and second-line RCC alone. And to be clear, this is a revenue opportunity to Arcus, not TAM or total addressable market. There's opportunity to expand that further with trials that support casdatifan usage in other RCC settings as well as potentially other tumor types like HCC, which Terry referenced earlier. And as a reminder, we own all of the rights to casdatifan, including economics, other than in Japan and certain other Southeast Asian countries, so nearly 100% of casdatifan revenues would accrue to us. I'll end with one slide that supports our plans to focus on a TKI-free regimen in the front line, and that's Slide 24. Here, we show market research that we recently conducted to determine the preferred combination partner for CAS in the front line. You can see here very clearly that the IO/IO regimen is preferred 3x more than the IO/TKI regimen as a combination partner for CAS for all the reasons we discussed earlier today. I'd now like to turn the call over to Jaen to discuss our immunology portfolio in more detail.

Thanks, Jen. I'm happy to describe such emerging immunology portfolio. This is an area in which we've had interest and in-house expertise since our -- the funding of the company. This portfolio currently includes 5 programs that you can see in our slide deck, reflecting the maturity of multiple years of research and compound optimization. Our strategy is simple and has been core to our approach since the very beginning of the company, to minimize biological risk by focusing on and building upon mechanisms with validated clinical activity. In general, we are focused on taking a small molecule approach to pathways where biologics have been highly successful. This is where we believe that we have an opportunity to create the most value. We generally look for more than just the convenience of an oral drug. In fact, the value proposition for several of our programs includes an expectation of differentiated efficacy, safety relative to the approved biologics. Our portfolio includes several programs, as I mentioned before, 5 active ones. But today, I will focus on our MRGPRX2 and TNF inhibitor programs, which are expected to be the first to reach the clinic. Firstly, regarding our X2 antagonist program, we plan to target both chronic urticaria as well as atopic dermatitis. X2, which has received a lot of attention in recent months, represents a novel and exciting target for the modulation of mast cell activity. While anti-IgE and anti-IL-4 receptor antibodies are marketed very successfully, for example, Dupixent alone is a $15 billion drug, there's still significant unmet medical need. While with the marketed antibodies, a substantial percentage of patients don't respond well to treatment. I want to emphasize that targeting mast cells is a validated approach in allergic conditions such as chronic urticaria and allergic asthma. A small molecular approach that targets a novel pathway for mast cell degranulation could represent a differentiated mechanism of action to address the existing clinical need, both in chronic urticaria as well as atopic dermatitis relative to the available biologics. Our candidate molecule has been specifically designed to improve both on the potency and pharmacokinetics side of the equation relative to the early small molecule entrants into the clinic. In fact, we believe that our required clinical exposure in patients, obviously, will be dramatically lower than those required for the leading small molecule currently being evaluated in clinical trials. This could translate into a potentially improved therapeutic index and potential best-in-class profile. We expect our MRGPRX2 antagonist to enter the clinic later this year. We plan to start with a healthy volunteer Phase I study and to follow quickly with a proof-of-concept study in chronic inducible urticaria. This program has the potential to generate proof-of-concept data within 9 to 12 months following entry into the clinic. Our TNF inhibitor program also represents another significant opportunity. Anti-TNF antibodies are amongst the most successful drugs ever developed. With our small molecule approach, we have the potential to develop a Humira in a pill. One of the challenges with today's TNF antibody therapies is that they block TNF signaling at both receptor 1 and receptor 2. However, blocking TNF receptor 2 can actually impact regulatory T cells and tissue repair, resulting in a paradoxical inflammation as a side effect in some patients. Our approach is designed to selectively prevent TNF from activating TNF receptor 1 while preserving TNF receptor 2 biology, which we expect to be an effective but safer alternative to the antibodies. Relative to the early competitor in the clinic, our molecule has been designed to have a better overall potency and human PK profile, and we expect to be in the clinic with this program in late '26 or early 2027. Similarly to the X2 program, the TNF program also has the potential to generate proof-of-concept data fairly quickly. We are very excited about the potential for our immunology programs to provide improved options for patients, and we are working to move these into the clinic as rapidly as possible. With that, I'd like now to hand it over to Bob to review our financials.

Thanks, Juan. Our cash at the end of the fourth quarter was $1 billion as compared to $841 million as of the end of the third quarter. Our cash position was bolstered by proceeds from our $288 million financing in November. Turning to our financial results for the quarter. We recognized GAAP revenue for the fourth quarter of $33 million, which compares to $26 million for the third quarter. Our revenue is primarily driven by our collaboration with Gilead. Our R&D expenses for the fourth quarter were $121 million as compared to $141 million in the third quarter. G&A expenses were $26 million for the fourth quarter compared to $27 million for the third quarter. Total noncash stock-based compensation was $15 million for the fourth quarter compared to $14 million for the third quarter. Shifting gears to guidance for 2026. We expect to recognize GAAP revenue of $45 million to $55 million for the full year of 2026. As we discussed on prior calls, we expect operating expenses to decrease meaningfully in 2026 as compared to 2025. The magnitude of this decrease will be in part determined by the results of the futility analysis for STAR-121, which will be conducted in the next couple of months. Accordingly, we expect to provide more detailed R&D expense guidance in connection with our Q1 call. We expect our cash and investments will enable us to fund operations until at least the second half of 2028. For more details regarding our financial results, please refer to our earnings press release from earlier today and our 10-K. I'll now turn it back over to Terry.

Thanks very much, Bob. I'm going to end on Slide 33, spend a bit of time on our upcoming news flow for 2026. As I mentioned at the beginning of the call, this is going to be another huge year for casdatifan data. Those data are both going to reinforce the confidence in its best-in-class profile and provide greater clarity on our first-line development strategy. Later this year, we'll have at least 2 additional data presentations for CAS. First, we'll be presenting updated data for the cas plus cabo cohort. We'll do that either at an investor event or medical meeting, by which time we'll have a minimum of 12 months follow-up on all patients. So the aim here is to be able to provide a relatively mature and meaningful data set, including Kaplan-Meier curves. Also later this year, we'll share new data from the casdatifan plus zim cohort of ARC-20. This is intended to demonstrate the safety and early efficacy of the combination and to establish CAS plus anti-PD-1 as the backbone for our first-line combination. There's also a lot happening on the development front. For PEAK-1, our first Phase III study for casdatifan, our goal is to complete enrollment by year-end. Also, we're adding first-line combination cohorts to ARC-20 to determine the optimal regimen for our first Phase III study in the frontline setting, and this will enable us to initiate our second Phase III study for casdatifan at the end of this year. We also expect to advance our 2 lead programs targeting inflammation and autoimmune disease into the clinic by early next year. With that, I'd like to thank everybody for joining us. We appreciate your interest, your continued support of Arcus, and we're happy to open the call for questions.

[Operator Instructions] Our first question comes from Salim Syed from Mizuho.

Congrats on the data at ASCO GU. Just one from us on -- and I appreciate all the color around the first-line strategy. Just the numbers that you provided here on PD and PFS, are there any particular trials that you guys are looking at just so we can start to think about the right IO/IO benchmarks here as you move away from TKI-based regimens? And just second to that, any thoughts here on what you would like to see in terms of improvement from those benchmarks?

Sure. So that's a good question. And I think as we move into this TKI-sparing regimen, the focus should really go to ipi/nivo. So there's a lot of data out there on ipi/nivo. Those of you who were at ASCO last year know that there were even like basically a decade of follow-up. And the reason is ipi/nivo is the #1 therapy that is utilized in the front line. And it's roughly in 1/3 to 35% of patients. And one of the reasons -- it is growing, but one of the reasons that it isn't used even more is the rate of primary progression, and that's on the order of 20% to 25%. Second, it has a relatively short PFS. It's on the order of 12 months or so. So I think those numbers become very clear, simple lines to put in the sand where, obviously, we want to demonstrate meaningful improvement. And I would emphasize that's all going to start with that rate of primary progression. And we believe it's exciting that we've even shown with anti-PD-1 alone, we're looking at potentially single-digit rate of primary progression, and we'll have more to say on that later this year. But I think those form a good place to start the conversation around benchmark.

Yes. I just would like a specific study. So like CheckMate-214, which was a registrational study for ipi/nivo in frontline RCC. And then the other study to look at that had an ipi/nivo arm with COSMIC-313. And interestingly, the key efficacy measures. So PD, PFS, OS, et cetera, were all very, very similar across those 2 studies.

Our next question comes from Li Watsek from Cantor.

This is Daniel Bronder on for Li. Congrats on the data update as well from our end. We're just curious to hear where you're at with the volru plus cas run-in in the frontline study that's being operationalized by Astra

Yes. So thanks. So what the question is referring to is the collaboration that we have with Astra

Our next question comes from Daina Graybosch from Leerink Partners.

Bill on for Dana. Congrats on all the data. It looks really great. So in your biomarker analyses, you mentioned that the deeper EPO reductions were correlated responses. Is this due to higher baseline EPO? I guess, in other words, did patients generally reach a similar, I guess, numeric EPO level after treatment? And I have one follow-up.

So I'll let Juan give you some comments on that.

Yes. So there's a -- what I would describe as a soft correlation between baseline levels being higher and the extent of reduction. Both of those we interpret we have data to support that those reflect the extent of HIF-2 alpha activity in those tumors. The tumors that have a stronger HIF-2 alpha signal tend to present with higher EPO levels, and those also tend to be the patients that display the deepest and more sustained reductions.

Got it. And when you're looking at these biomarkers, did you also look at VEGF and other genes that might also be downstream of HIF-2?

Yes. And that there are other soluble peripheral markers that we'll be disclosing later this year that are regulated by HIF-2 alpha and are well known to be each one of them independently negative prognostics in this setting.

One thing I would emphasize also about the biomarker work that I think is important for people to recognize, this is most important in just sort of supporting the overall understanding and tie to mechanism of the activity. But in this particular case, there's no sense that we would ever think about having or needing a selection criteria based on the biomarker. Because what I'll remind you is that probably 80-plus percent to 90-some-odd percent of patients with clear cell RCC have some level of HIF-2 as a driver. So when we report on these type of data, what you're actually seeing is while there's correlation, even those patients with lower levels are -- there is benefit. So we get a pretty continuous spectrum of benefit throughout those patients as opposed to some sort of binary cutoff that you might be seeking.

Our next question comes from Richard Law from Goldman Sachs.

This is [ Jane ] on for Rich. Congrats on the progress. So we have 2 questions about the upcoming presentation at ASCO GU this weekend. So first, Merck will present detailed results of belz plus lenva in the Phase III LITESPARK-011, which may set the bar in second line. So what are your thoughts and expectations for LITESPARK-011? And my second question is, so Merck is also running a study of belz plus zanza in the Phase Ib/II KEYMAKER-U03. So they have a poster about TIP poster this weekend. So what's your view on this combo strategy versus cas plus cabo?

So let me start with expectations around LITESPARK-011. From everything that Merck put out, they obviously said that the study was successful and any body language that you might infer, -- we think the data are going to look quite good. And we're excited about that. We think a couple of things. First off, it's really providing important validation for the field in an earlier line setting. And we feel great knowing that we have a better HIF-2 inhibitor, and we're going on top of what is the standard of care cabo in so far as TKI versus lenva, and we have the same control arm. So from our perspective, obviously we have no knowledge of anything quantitative, but we expect good data and the better the Merck data, honestly, the better we feel it will be better for patients will be better for Arcus because we think we're going to outperform them with both of the molecules in our combination. The second part, when I say there's 2 important things, is all of our investigators have emphasized this point which we recognize. We have a lot of tailwinds in terms of enrolling PEAK-1. But for a number of reasons, the positive data that they'll present is going to really help us to drive enrollment. As I mentioned, we're looking to be fully enrolled by the end of this year, and we think the LITESPARK data will help us very much. I'll make a couple of comments on zanza, and I'll see if Jen or Richard want to add anything more on top of it. But we just simply don't see zanza as being a key changer, at least in clear cell RCC at this point. So we think cabo is the clear standard of care. It's -- not only does it have a great profile, but the reality is it's very entrenched. So it's used a lot. Clinicians know how to use it. They know how to deal with the AEs. They're very comfortable with it. So we think cabo is the TKI of choice.

Just one note on KEYMAKER-U03 you were asking for. The presentation is just a tip from the KEYMAKER study, which is just a Phase Ib/II platform study. So there's no data that's going to be presented for belz. It's a trial in progress poster.

Our next question comes from Asthika Goonewardene from Truist.

This is [ Cardi ] on for Asthika from Truist. Just on the first question on ARC-20 monotherapy, the last update had ORR somewhere in the 30% range, and now you're showing improvement in the mid-40s. So how much of that was due to deepening responses? And at what time point were you seeing responses start to deepen?

So it's definitely all due to deepening of response. And I like the question because it does go back. We share a lot of data and we share almost in real time. And so one of the slides we've included, we've shown how those data have evolved over time. But we also pointed out at the earliest time point, how with this mechanism and also the safety profile, it's very clear that responses can occur even out past a year. So the responses occur at all sorts of different time points. The stable disease patients also continue to do very well. The question about DOR hasn't come up. But what I can say is we're not close to a DOR. So patients, once they can get past that initial scan, do very well, and that's why we're seeing the PFS we're talking about. But there's no generalities in terms of where those responses may occur. And what I would emphasize to your question on deepening, even when we've seen patients that have generated their initial response out past a year, it's meaningful. So it isn't just that they were at a 29.6% tumor reduction and they went to 31%. These are genuine deepening of response. And the way we interpret it is that I think people aren't used to looking at oncology mechanisms that have a relatively benign safety profile. And what I mean by that is they're not poisoning you while they're reducing the size of your tumor. So once these patients become stabilized and they become healthier and they become stronger, all those things kick in, including their own immune system, to help. And so that's why you may see patients that all of a sudden deepen, it may have not even always been a gradual deepening, they start to deepen later on. And you do have for this mechanism, what looks very IO-like in terms of the tail. So patients that do well, do well and they do well for a long time. And that's one of the reasons we feel and I think justified by the data that the market is not only going to be driven by the number of patients, but by the durability of the treatment. And we've had a lot of patients, even a single agent in the late line out past 2 years, we think that the front line is going to be something where you can have patients going 3, 4, 5 years, and that gets the whole TKI sparing strategy that you're not just pushing off that TKI for a matter of months, but you're going to dramatically change the opportunity for that patient that first presents in so far as not having to go for TKI until many years later. But they'll still ultimately get whatever benefit you might get from the TKI, but you just flip around the paradigm of no longer requiring that at the outset to get the tumor under control, but they can get that years into their therapy.

Got it. If I may, I'm just going to squeeze in one more question.

Sure.

Okay. So you have said that STAR-121 will have a fertility analysis in the coming months. So if you choose to discontinue STAR-121 based on the results, what is the clinical impact? And what is the impact on your R&D spend?

So the clinical impact is we're already anticipating from an operational standpoint that that likely will occur, but we may see something that tells us to keep going. Impact-wise, operationally is fairly minimal because the study is basically fully enrolled. So that impact operationally would be that you won't be doing all that work leading up to the registration. I'll let Bob comment on how we think about sort of the expense part of that.

Yes. I mean the vast majority of the expense that we see for especially any of our late-stage clinical trials really is incurred primarily through the enrollment cycle and as the primary portion of patient treatment is occurring. So when you get to the latter parts of the trial, and you've presumably seen that decrease in expense as an example, for 221 and 121 and other studies, when we get to the latter part of the life cycle of trials, the expense starts to drop off pretty markedly.

Our next question comes from Yigal Nochomovitz from Citigroup.

I just want to probe a little bit more in terms of the frontline strategy. So you've indicated that cas plus zim will be the backbone of this strategy. And then the question, of course, is whether the CTLA-4 comes into the mix. So with that in mind, I'm just curious, when you asked the oncologist, as you showed us this new market data, in that question, you didn't put in the question on cas plus zim. You put the triple. So I'm just wondering is the interpretation there that your base case is really to do PD-1, CTLA-4 CAS and that's where it's going to shake out? Or is there some reasonable potential that it could end up just being what you've identified as the backbone, cas plus zim?

I'll let Jen start on that. We'll see if I add anything on top of what she said.

Yes, I mean, I think you actually described it really well that right now, our base case assumption is that it would be cas plus ipi/nivo. And so right now, we're thinking of cas plus zim or cas plus anti-PD-1 really as a backbone and that we would likely build on that. We'll keep looking at the cas plus zim data the outlook over time. But yes, certainly, base case right now is cash plus ipi/nivo. And as we talked about today, that cohort is actually now enrolling and the idea is to generate safety data very, very quickly. And again, an early look at efficacy by looking at the rate of primary progression. And so what's also nice about this combination, as Terry was talking about earlier, because ipi/nivo has a relatively high rate of primary progression, if we see early on at the first scan that that rate of primary progression is coming down, that gives us a very early read on efficacy, which is nice to have.

So what I'll add, Yigal, and this actually it's come -- questions come up from investors, and it's a topic also with investigators. So the data that we've seen thus far, and obviously, we just -- they were early and we showed you the primary progression, but it's a good start. And the question sometimes does come up in the context of casdatifan, how much value does the anti-CTLA-4 bring. So obviously, we're going to have an early data set. But the way we look at it is if you think about it, and this comes up, even though it wasn't a formal part of that analysis that Jen did, when you talk with investigators, including some of those who are aware of the anti-PD-1 early data more specifically, as you could imagine, anti-PD-1 plus CAS, if you could bring that into the front line from a patient standpoint, that would just be awesome. I mean to not have to go with anti-CTLA-4 and not have to go with TKI would be like it comes up sometimes like a vacation. So we are going to pay a lot of attention to that. With that said, I'd like to also put the other conservative part of this that we think is important. We don't want to get too cute. So if that anti-PD-1, anti-PD-1, anti-CTLA-4 each on top of CAS looks similar or maybe like anti-PD-1 potentially could be there, we're still more likely than not to go for a 3-arm study where we would have anti-PD-1, anti-CTLA-4 CAS versus anti-PD-1 cas versus ipi/nivo. So we're not going to get too cute and outsmart ourselves just based on early data. But if it should, we're going to give anti-PD-1 CAS, it's shot at the limelight, if possible, because it would be great for patients.

Our next question comes from Jonathan Miller from Evercore.

I'll follow up on Yigal's question on the first-line setting. You say at least one Phase III starting this year. You're adding an additional undisclosed combo to ARC-20 in the first-line setting this year. The ipi/nivo combo is open. You just said it was sort of your base case, but you haven't committed to showing data for us ahead of making that Phase III decision. So I just want to get a sense for broadly speaking, how many first-line Phase IIIs are you thinking about in aggregate eventually? What are your plans for the adjuvant setting, which I noticed you didn't mention except saying that CAS belongs in all lines of therapy? And what are your current thoughts on partnering in RCC and beyond for CAS and whether that unlocks additional bandwidth to do some of these late-stage design?

Yes. So I'm just going to comment briefly and then turn it over to Jen to give a more fulsome answer on all of that. But I'll address the partnering part. From a partnering, all you should expect to potentially see and you probably will see are like clinical collaboration. So at this point, we feel -- we love that we basically own 100% of the casdatifan rights other than in Japan and a few other Southeast Asian countries. As you know, that gives us an enormous strategic optionality. With that said, there are other mechanisms and settings that make sense. And since the casdatifan is a relatively rare beast, we feel like we're in a good position to do some smart clinical collaborations under good terms. I'll let Jen comment more broadly, though, on the number of Phase IIIs we might do, how they're sequencing, when we'll disclose some data, et cetera, because we will obviously -- we won't make a decision without sort of giving a sense of what drove that decision. But go ahead, Jen.

Yes. So just on the Phase III front, obviously, we have TKI-1 that's enrolling. Our plan is to start one other Phase III study around year-end. That would be in the frontline setting and informed by this new cohort that we're adding that you referred to, Johnson, a cohort that's looking at CAS plus anti-PD-1 plus CTLA-4. So you could probably make an assumption that that first Phase III study would be looking at that triplet versus ipi/nivo. So that would probably be our base case assumption today. We'll probably also add, as you were pointing out, another combination to ARC-20 to look at CAS plus anti-PD-1 plus another mechanism, probably something that wouldn't be a big surprise to a lot of people. I'd say as far as whether we took that into a Phase III is very TBD right now. I think right now, that first Phase III in the frontline setting that I mentioned is our top priority and what we'd really be focusing on from a resource perspective. But we're also interested in generating some other data with that other combination as well. And then on the adjuvant setting, we'll see what the LITESPARK-022 data looks like. And I think right now, we probably view that as a lower priority relative to certainly frontline and maybe some other things we might do with CAS, including HCC, just given it's not a huge market, patients are on treatment for 12 months max. It's a very high bar from a safety perspective in the adjuvant setting just because these patients are doing pretty well and feeling well. So a lot of times, they don't want to be on therapy, they come off therapy. So like I said, we'll see their data. We're always evaluating it. We'll continue to evaluate and something that we might consider for the future.

I think the way -- and obviously, as the year goes along, we'll continue to share and probably into next year, we'll be sharing. Like our view is that HIF-2 inhibition is going to be used in every line in every setting of clear cell RCC. And our ultimate development strategy whether it's supportive studies, et cetera, is going to look like we're going to make sure we've covered everything with time. But from a prioritization standpoint, it's PEAK-1 frontline, potentially another one next year. And then we'll be surrounding those with other studies that make sense so that we're leaving no stone unturned in being able to be used and reimbursed, et cetera, in every line of therapy.

Yes. And we're continuing to look at clinical collaborations for that other new frontline option that we're thinking about. So not everything is on our dime. It's not all our resources, et cetera. So that's important to us as well.

Great. And if I could just squeeze in one more before we run out of time. On PRX2, you've mentioned a couple of times the potential for a safety delta on the basis of better potency and lower dosing. But are there specific safety signals that we should be looking at when we think about initial data here? And how much data from the initial cohorts, especially in healthy volunteers, will you need to be able to put some bookends around what that potential safety delta might look like?

Right. So I'm always saying that outside of oncology, you always one dose away from complete disaster. So the answer is that a well-run healthy volunteer study will give you some comfort, but you're always accumulating additional data. The thing to -- the default thing to always look out for, and I think that some of our competitors in the space sort of generate a little bit of a hint of this, when you're dealing with a very high exposure of any xenobiotic, you need to look at liver function, okay? And so you can go -- you can actually -- as our competitors have seen, if you go too high, you will actually start to see the liver complaining. And so we think that we'll be able to put way more daylight between our -- the amount of our drug required to elicit a similar pharmacology and levels where the liver is going to start to complain.

Our next question -- our last question, sorry, comes from Emily Bodnar from H.C. Wainwright.

I was going to ask if you could give us some updated expectations for the cas plus cabo data later this year, given now you have pretty mature monotherapy data with PFS of at least 12 months. So how are you kind of thinking about data for the full 45 patient data set? Hello?

There seems to be a connection issue. We'll be right back in just a... Emily, could you please repeat your question?

Yes. Do you hear me?

Yes. Thank you, Emily.

Okay. Yes. I was going to ask if you could give us some more updated expectation through your cas plus cabo data later this year, given you now have pretty mature monotherapy PFS is over 12 months. So how are you kind of thinking about what the combo could look like in earlier line patients in that full 45 patient data?

Jen?

Yes. So as Terry said, when we present the data, the goal is to have 12 months minimum follow-up on everybody so that we may not have a PFS, but we'll be able to look at a Kaplan-Meier curve and at least make some educated guesses as far as where PFS could shape out or shape up. And so as you know, for cas mono PFS, we're seeing a range of 12 to 15 months. So we'll see what cas cabo shows. And right now, we obviously believe very, very strongly that we can be cabo alone because cas mono alone looks better than cabo. And so we'll build on both what cas mono looks like and what cabo mono looks like. So yes, so we're excited to get those data out. And we think like the LITESPARK-011 data, it will also be very derisking for the PEAK-1 study.

The other nice thing about LITESPARK-011, just may be obvious, but it will give a good, call it, contemporary look at what cabo alone looks like because that's the control arm. So to the extent that thinking about benchmarks.

Yes. And what gives us obviously a lot of confidence in the cas plus cabo data in the PFS is, like I said, just what we're seeing with cas mono and the fact that cas mono looks better than cabo mono and we think may even look better than belz plus lenva, you'll see what the LITESPARK-011 data shows.